B4PythonIntermediateAbout 120 min to complete
Random Numbers
Reproducible randomness, and the ability to turn uncertainty into a number you can defend.
Notebook
About 120 minutes to complete
Enrol to read the notebook and download it to run.
What this module covers
Reproducible randomness, and the ability to turn uncertainty into a number you can defend.
By the end of it
- Use the modern default_rng Generator API and explain why the legacy API is discouraged
- Seed generators correctly, and state exactly what reproducibility guarantees you have
- Generate values from the distributions you will actually meet in data work
- Choose the right distribution for a given real-world quantity
- Sample with and without replacement, including weighted sampling
- Shuffle and permute data safely for train/test splits
- Build a Monte Carlo simulation and read a probability off the result
- Bootstrap a confidence interval without any distributional assumptions